Re: [PATCH] maint.mk: catch "see @xref{}"

[Jim Meyering]

Akim Demaille wrote:

> Le 16 avr. 2012 à 19:19, Jim Meyering a écrit :
>
>> Good one.
>> I prefer to use " +" rather than "  *",
>
> You are right, I started for grep, not grep -E.
>
>> and suppose someone might say "See also ...", too,
>> so added the (also +)? term.
>>
>> Do you think it's worth worrying about "Also see @xref...", too?
>
> which includes "see @xref" :)
>
>>
>>
>> diff --git a/top/maint.mk b/top/maint.mk
>> index 2228a37..700dd1c 100644
>> --- a/top/maint.mk
>> +++ b/top/maint.mk
>> @@ -838,6 +838,13 @@ sc_texinfo_acronym:
>>      halt='found use of Texinfo @acronym{}'                          \
>>        $(_sc_search_regexp)
>>
>> +# Avoid double "see" when contributed by @xref.
>> +sc_texinfo_see_xref:
>> +    @prohibit='[Ss]ee +(also +)address@hidden'                              
>> \
>> +    in_vc_files='$(texinfo_suffix_re_)'                             \
>> +    halt='found use of "See @xref{}"'                               \
>> +      $(_sc_search_regexp)
>
> That's good with me.  I would be happy to be able to catch broken
> lines though.  It would be nice to be able to escape from the constraint

Hey!
We can/should use perl, and there's even an existing framework for this.
Change the default prohibit_undesirable_word_seq_RE_ to e.g.,

- /\bcan\s+not\b/gims
+ /\b(can\s+not\b|(see(\s+also)?|also\s+see)address@hidden)/gims

> of grep per-line.  In Texinfo "\n\n" is a better separator imho.  How
> about using Perl instead of Grep?

Nice!  That new rule exposes problems in coreutils.texi:

  doc/coreutils.texi:4913:see @xref{
  doc/coreutils.texi:5051:see @xref{
  doc/coreutils.texi:16003:Also see @xref{
  doc/coreutils.texi:16112:see @xref{
  maint.mk: undesirable word sequence

So how about this, instead?

>From 734965bbccf994b40f844c933d8f7fe4213b4ef6 Mon Sep 17 00:00:00 2001
From: Akim Demaille <address@hidden>
Date: Mon, 16 Apr 2012 20:21:51 +0200
Subject: [PATCH] maint.mk: catch "see @xref{}" and similar

* top/maint.mk (prohibit_undesirable_word_seq_RE_): Also
prohibit "See also @xref{", "Also see @pxref{", and similar.
---
 ChangeLog    |    7 +++++++
 top/maint.mk |    5 ++++-
 2 files changed, 11 insertions(+), 1 deletion(-)

diff --git a/ChangeLog b/ChangeLog
index c08ba76..479c360 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+2012-04-16  Akim Demaille  <address@hidden>
+       and Jim Meyering  <address@hidden>
+
+       maint.mk: catch "see @xref{}" and similar
+       * top/maint.mk (prohibit_undesirable_word_seq_RE_): Also
+       prohibit "See also @xref{", "Also see @pxref{", and similar.
+
 2012-04-16  Jim Meyering  <address@hidden>

        bootstrap: really use gnulib's po/Makefile.in.in
diff --git a/top/maint.mk b/top/maint.mk
index 2228a37..706d8fe 100644
--- a/top/maint.mk
+++ b/top/maint.mk
@@ -932,8 +932,11 @@ sc_prohibit_doubled_word:
 # A regular expression matching undesirable combinations of words like
 # "can not"; this matches them even when the two words appear on different
 # lines, but not when there is an intervening delimiter like "#" or "*".
+# Similarly undesirable, "See @xref{...}" expands to "See *Note ...".
+# In general, remove a preceding "see".  I.e., if you use "Also see @xref{",
+# change that to "Also @xref{".
 prohibit_undesirable_word_seq_RE_ ?=                                   \
-  /\bcan\s+not\b/gims
+  /\b(can\s+not\b|(see(\s+also)?|also\s+see)address@hidden)/gims
 prohibit_undesirable_word_seq_ =                                       \
     -e 'while ($(prohibit_undesirable_word_seq_RE_))'                  \
     $(perl_filename_lineno_text_)
--
1.7.10.169.g146fe